Our understanding of the causation of brain tumors has been advancing slowly, because there are so few epidemiologists interested in the investigation of these diseases. My training in both clinical epidemiology and neuro-oncology has prepared me well for a career in brain tumor epidemiology. My immediate goals are to obtain further research training in brain tumor epidemiology, genetic epidemiology, cancer chemoprevention and population-based research through one-on-one mentoring, didactic lectures, seminars, meetings and hand-on experiences. In the next 5 years, I will seek opportunities to build a network of collaborators and to conduct pilot association studies to evaluate susceptibility factors in brain tumors. Initially, my research will take place at the New York Presbyterian Hospital, but after I establish a working relationship with the New Jersey Cancer Registry and the Brain Tumor Epidemiology Consortium, I will apply for funding to initiate population-based investigations and to collaborate with other epidemiologists in multicenter studies. My long term goal is to independently obtain NIH funding to examine the genetic susceptibility of brain tumors and to expand the potential of chemoprevention in these devastating diseases. Scientific Plan: I will recruit 200 cases and 400 frequency-matched controls at New York Presbyterian Hospital to evaluate the association between statin therapy and risk of malignant glioma. Based on these drugs' inhibition of the HMG-CoA reductase pathway, their immunomodulatory, anti-inflammatory and potentially anti-angiogenic effects, there may be an inverse association between the use of statins and risk of malignant glioma. In order to identify a genetic subset of individuals most likely to benefit from statins, I will also evaluate haplotype-tagged single nucleotide polymorphisms (htSNPs) of genes in the HMG-CoA reductase pathway, Rho and Ras GTPases as modifying factors for statins therapy. This strategy will allow the selection of those individuals with the right genetic profiles as drug candidates. Relevance: This will be one of the first studies that evaluates a chemopreventive approach in malignant [unreadable] glioma, and as such, will help to open up a new therapeutic approach in these diseases. Using an individual's genetic makeup to predict response to treatment may improve the cost effectiveness of using these drugs. [unreadable] [unreadable] [unreadable]